Wunderman Thompson uses data and AI to create relevant experiences for brands
Merging creativity, data and technology
Using data to shape new messaging or find new prospects is core to our business, but we wanted to do more, and do it better. In markets continually roiled by disruption and innovation, we needed to help our clients move beyond transactional relationships toward cultivating deeper, longer engagements, using data to forge authentic interactions between brands and customers.
Our ultimate goal at Wunderman Thompson was to build our machine learning and AI capability so we could create more accurate models and scale that capability across the organization. We had implemented some machine learning, but siloed databases constrained our ability to use predictive modeling effectively. To tune our operations for AI, we needed to dissolve the silos, merge the data and infuse it across the business. We needed to build a unified data science platform, a single data ecosystem that could serve our organization and beyond.
Our largest databases — iBehavior Data Cooperative, AmeriLINK Consumer Database and Zipline Data onboarding and activation platform — the most extensive in the industry, comprise billions of data points across demographic, transactional data, health, behavioral and client domains. Combining these properties would provide the foundation to instill machine learning and AI across the business.
How could we transform our data practice, fully integrating machine learning into the business? Make our data ready for AI in a hybrid cloud environment? We needed a robust platform, an open information architecture, that would allow us to maximize and consolidate our assets in a multicloud environment.
Enlisting the IBM Data Science and AI Elite team
To resolve this multifaceted challenge, only expert help from a trusted provider with innovative technology, industry expertise and enterprise ready capabilities would do. A long history of working with IBM led us to IBM Analytics, the Data Science and AI Elite team.
With the help of IBM’s Data and AI Expert Labs and the Data Science and AI Elite team, we built a pipeline that allowed us to import the data from all our three largest data sources. This combined asset contains more than 10TB of data amassed over more than 30 years from hundreds of primary sources, including precise data for more than 260 million individuals by age; more than 250 million by ethnicity, language and religion; more than 120 million mature customers; 40 million families with children; and 90 million charitable donors.
With the ability to work collaboratively across many different regions and offices, we were able to run models in a way that previously had been impossible. When the Data Science and AI Elite team introduced us to AutoAI, that’s when the work really scaled up.
John Thomas, IBM Distinguished Engineer and Director, IBM Analytics, led the creation of a system that combined Watson Studio and Watson Machine Learning. With AutoAI as the linchpin, we created an automated end-to-end pipeline to bring as much information as possible into our data pool, delivering more data to fuel better predictions, generating better prospects for clients. We used IBM Watson Studio to support model building and prediction and developed an iterative model selection and training process until the models met the appropriate criteria.
Eight weeks of collaboration with the Data Science and AI Elite team and industry insights from the IBM Account team delivered a proof-of-concept, undergirded with a sound methodology that enabled better-performing models using enriched datasets. We compared data points in each source to filter out records for desired features and reconciled these against one another. We subsampled tens of thousands of records for feature engineering, applying decision tree modeling to highlight and select the most important features for training data.
The results showed a significant uplift over previous models, a dramatic increase in segmentation depth, raising rates well beyond our initial projections. With an average change from 0.56 to 1.44 percent, a boost of more than 150 percent, IBM helped us uncover new personas in existing databases we’d previously been unable to reveal, delivering a dramatic improvement in deliverable customer lists.
Confidence and capability to make precise predictions
The ability to use all of our data, to use these techniques, with human insight and understanding, really gives us a best in class capability to find new customers for any of the brands we serve. And it includes all of the data, at full scale, with much more advanced machine learning and the ability to run all of that processing on elastic compute inside various cloud providers. There’s a capacity here that’s frankly market moving.
We now have the confidence that we can more accurately predict which customers will respond to campaigns, that we can find new audiences based on correlations to existing customers.
This new machine learning and AI solution delivers the power to personalize messaging at scale to create meaningful, more resilient relationships with more customers – meeting their needs no matter what circumstances the world is facing. And that allows us to build more revenue for our clients, and for our business.
Learn how the IBM Data Science and AI Elite team can help you harness data science and AI to bring value to all aspects of your business.